Prediction of type 2 diabetes using simple measures of insulin resistance - Combined results from the San Antonio Heart Study, the Mexico City Diabetes Study, and the Insulin Resistance Atherosclerosis Study
Prediction of type 2 diabetes using simple measures of insulin resistance - Combined results from the San Antonio Heart Study, the Mexico City Diabetes Study, and the Insulin Resistance Atherosclerosis Study
复制标题
DOI：
10.2337/diabetes.52.2.463
复制
发表时间：
2003-02-01
期刊：
影响因子：
7.7
通讯作者：
Haffner, SM
中科院分区：
文献类型：
--
作者：
Hanley, AJG;Williams, K;Haffner, SM
为了确定并正式比较胰岛素抵抗（IR）简单指标预测2型糖尿病的能力，我们使用了来自圣安东尼奥心脏研究、墨西哥城糖尿病研究和胰岛素抵抗动脉粥样硬化研究的综合前瞻性数据，其中包括非西班牙裔白色、非裔美国人、西班牙裔美国人和墨西哥受试者的良好特征队列，并进行了5-8年的随访。Poisson回归用于评估每个候选指标预测随访检查时发生糖尿病的能力（3，574例受试者中有343例发生糖尿病）。计算每个指标的受试者工作特征（AROC）曲线下面积并进行统计学比较。在汇总分析中，Gutt et al. 0和120 min时的胰岛素敏感性指数（ISI 0，120）显示最大的AROC（78.5%）。该指数比AROC曲线在66 - 74%之间的一大组指数（包括Belflore、阿维尼翁、Katz和Stumvoll的指数）具有更高的预测性（P < 0.0001）。在调整协变量后以及按葡萄糖耐量状态和种族/研究亚组单独进行分析时，这些结果基本相似。总之，我们发现在预测糖尿病方面，已发表的IR指标之间存在显著差异，ISI 0，120始终显示出最强的预测力。瘦指数可能反映了糖尿病发病机制的其他方面，除了IR，这可能解释其强大的预测能力，尽管其中度相关的IR的直接措施。
To determine and formally compare the ability of simple indexes of insulin resistance (IR) to predict type 2 diabetes, we used combined prospective data from the San Antonio Heart Study, the Mexico City Diabetes Study, and the Insulin Resistance Atherosclerosis Study, which include well-characterized cohorts of non-Hispanic white, African-American, Hispanic American, and Mexican subjects with 5-8 years of follow-up. Poisson regression was used to assess the ability of each candidate index to predict incident diabetes at the follow-up examination (343 of 3,574 subjects developed diabetes). The areas under the receiver operator characteristic (AROC) curves for each index were calculated and statistically compared. In pooled analysis, Gutt et al.'s insulin sensitivity index at 0 and 120 min (ISI0,120) displayed the largest AROC (78.5%). This index was significantly more predictive (P < 0.0001) than a large group of indexes (including those by Belflore, Avignon, Katz, and Stumvoll) that had AROC curves between 66 and 74%. These findings were essentially similar both after adjustment for covariates and when analyses were conducted separately by glucose tolerance status and ethnicity/study subgroups. In conclusion, we found substantial differences between published IR indexes in the prediction of diabetes, with ISI0,120 consistently showing the strongest prediction. Thin index may reflect other aspects of diabetes pathogenesis in addition to IR, which might explain its strong predictive abilities despite its moderate correlation with direct measures of IR.